Recently, I heard of several approaches to mastering a track... One is treating each section or part separately, so quieter parts get a dedicated and proportional treatment, with automation of some form. Another is dealing with the heaviest part and having that broad brush coat the whole thing for cohesion. Yet another would be finding an average that supports both lighter and heavier parts (also static, like the previous one). What is the most common way of doing things? Is automation all too common during mastering, or is it overcomplicating things? I've automated so much during mixing and was hoping to escape that at the mastering stage.

 

About "overcomplicating things", it's ironic, but many of the times I see someone mention this; they will explain all the plugins they might need to use to fix something or the option of just using automation. Almost like they are asking for someone elses' permission to do it.

 

when you think of mastering a whole album this might shed some new light on the topic.

 

This is the kind of question I’d love to hear answered by Dale Becker or Mandy Parnell, haha. Personally, I don’t think it’s very common, and if it does happen, it’s probably to fix a problem that couldn’t be solved any other way. But I’m not a mastering engineer, so I’m curious too

 

Last edited: Jul 24, 2025 at 6:09 AM

Time for another thread!

 

I wouldn't say that it's extremely common practice, but automating multiple different things on a master can provide minute changes or fixes to problematic mixes if necessary. It can also be a creative decision, but that's not typically how I've used it.

I have no problem doing minor volume automation if a track simply has sections that are a bit too quiet compared to the chorus/drop/high point. You could even sneakily automate the pre-chorus down by a dB and then bump it back up a dB when the chorus hits for a bit more impact. This is something that is commonly done during mixing for energetic rock tracks (chorus 1dB louder trick).

I've also automated the threshold for dynamic EQ bell cuts to deal with harshness on thing like claps and percussive high hats during mastering when they havent been dealt with correctly during mixing, and my processing on the rest of the track suits the mix well, but might bring out harshness in other parts. Once again I do this to a minor degree.

One thing to consider is that, if your already doing quite a bit of limiting on a track, volume automating a section up 1-2dB will basically result in you hitting the limiter harder, so you're going to be balancing quality degradation from more limiting with a small bump in perceived loudness.

I've also done some EQ automation for things like high and low shelfs to feature certain elements like a bass part, or add a bit of brightness to sections to make them more exciting. This is basically using automation to treat certain sections differently. These moves are typically small amounts for creative applications. For corrective applications I could be more aggressive.

Since moving to a more analog hardware focused setup, automating has become more tricky, but I have no problem adding in some digital tools for corrective automating!

 

There's really no reason why automation at the last step is suddenly not cool anymore, if it works, it works. No one will ever know.

The interesting part to me is that you say you already did a lot of automation in the mix, so that means you did the mix, and now you need to go and fix choices you did on the mixing stage. That seems backwards to me. Every time I master something that I mixed, when it gets to the mastering portion of the template, the sound is already pretty much locked in. Maybe this amount of automation is a symptom of a problem you need to fix in your mixes. Maybe you're obsessing over details that don't matter. Even worse if that's all while neglecting the basics.

If all you need is permission to automate the master, here you go: I, shinjiya, give you permission to automate your master.

 

This is a result of people talking about 2 different things the same way. When you talk about mastering a single song, inside the same DAW project as you did your mix; it's not the same work as someone mastering many songs together. You are done mixing before you are doing Wavelab audio montage work across multiple tracks, as an example.

 

What clone said is on the money.
There is more manual precision fine-tuning in mastering than automation. I have not seen any mastering engineer do it any other way except manually. Preset tried and true inserts saved and reloaded as starting points are not uncommon, but I wouldn't call it automation as much as I'd label it time saving.

 

Yes, the sound is locked in, but it's too quiet. Hence I need a limiter. Maybe a clipper, also. Hence issues pop up. I can always go back to the mix and recall details, no problems there. But at some point I'll have to say: this mix is done and I just need to make it loud to get it out. That last process has its own challenges. I haven't started yet and I'm just wondering whether it makes sense or not to automate (say a song has a quiet intro and a busy coda... think Stairway to Heaven).

 

Find out how they did it in the 80's & early 90's and just copy that.

 

This is true. It is common that professional mastering engineers might ride some knobs and faders while recording their analog mastering session. For example to add a little bit exitement to the chorus or vice versa lower some dynamics during verses. Same with intro/outro fades and filter movements. Some may be applied to the mix. But it's not uncommon that the decision how exactly a track starts and ends is done by mastering engineers. Especially in the context of an album.

Doing it manually is by definition not automation. But in this case I think it is used synonymous, because the intention is the same. Mastering engineers of course want your mix to sound cohesive and dynamically well thought, but on the other hand it is their job to exxagerate this for emotional impact. Automation (or "manually automate") is not a uncommon practice done by professionals as far as I know.

 

I dont know if it is common or not, but there's often some automation in my masters, if needed. Automating level or EQ or, more rarely, a dynamics parameter is very quick and bring better results than fiddling with plug-ins to find static 'one size fits all the tracks' settings. Automation can be recalling static settings at one point, not always dynamically changing parameters.
Mastering is always an album / EP / compilation thing. It doesnt take place in the void, it is always done specifically for this or that listening media. It is not track-oriented, it is an album/EP>that or this format thing. Mastering is where the bigger picture, the context in which the track will be listened to, is taken into account. Otherwise, it's mere stereo buss processing. Making one track sound good is a mixing job.

 

Last edited: Jul 24, 2025 at 3:36 PM

Talkin' about mastering is always so soothing...

 

As a mastering engineer I use automation on almost every track/song I recieve. It allows me to have full control of things - from dynamic progression to individual parameter modulation.

 

Automation in the Audio Mastering World
Let's look into a few practical examples of automation being used in audio mastering to achieve your desired mastered sound.
www.izotope.com/en/learn/automation-in-the-audio-mastering-world

 

The short answer is that there isn't a single parameter in the mastering chain that isn't potentially a target for automation if it's needed.

I've mastered albums where I didn't need to automate but also those that need a lot. The old adage of 'fix it in the mix, not at mastering' is certainly true but sometimes it's necessary. I used to do a lot of remasters of 60s and 70s albums and in those cases there was no going back to the mix to fix it (unfortunately in many cases). It's sometimes just easier to adjust something at mastering than rewinding the whole process back to the mix - it certainly was back when we were still dealing with multitrack tape.

 

Sounds like a great approach if you like a lack of consistency in one song

 

If the mix lacks consistency you could use automation to increase consistency between sections within the track. It really depends on what the tracks needs and your competence using automation.

There are definitely tracks that are not going to need automation, but some could really benefit from it.

 

I'd be careful with this kind of sarcasm without having a good argument.
You could easily archieve more "consistency" with the approach of automating parameters or riding faders.
In fact you could use it exactly for that matter.
Of course there are many other usecases.
Having the chorus more exiting than the verse is one of the most common thing in music.

What is your definition of "consistency" anyway?
Not sure I'm feeding the troll here.

[EDIT] The short and equally sarcastic answer is: you could easily fix the lack of consisteny the automation brought you by abusing clippers/limiters til your waveform is a straight brickwall.